A variety of genetic and biochemical studies have proved that virtually any biological process (i.e., cell behaviors and the like) can be broken down into components. This reductionist approach to biological inquiry seeks to understand the greater part of life's complexity in the relatively simple chemical terms of molecules and molecular interactions. In the middle part of the twentieth century, several scientists showed that metabolism can be understood as a series of enzymes that act sequentially to convert precursor compounds into final metabolic products. This insight gave rise to the notion of genetic or biochemical pathways that control cellular processes. More complicated cellular behaviors such as differentiation have recently been defined in terms of genetic programs and pathways. Even disease processes can be thought of in such terms. For example, the hypersensitive response of plants is a pathway characterized by cell collapse, cell dying, the deposition of callose, the physical thickening of cell walls by lignification, and the synthesis of various antibiotic small molecules and proteins. An effective strategy to study the hypersensitive response involves the elucidation of pathogenesis-response pathways.
Genes regulate some of the most commercially, agriculturally, and medically important processes in biology. However, determining which genes function in what pathway is a complex process. There are few methods available to screen large numbers of genes or promoters in plant cells for their effect on expression in a physiological pathway of interest. Co-bombardment methods that allow read-out of a marker-linked promoter based on the action of another gene in trans (See, U.S. Pat. No. 5,981,730, Ausubel et al (1995) Current Protocols in Molecular Biology, Greene Publishing and Wiley-Interscience, New York) are not amenable to high throughput analysis. Other methods that use Tobacco Mosaic Virus to screen for cDNA inserts that give a local lesion response in a normally non-necrotic mutant are not suitable for the detection of positive phenotypes. Methods such as the yeast two hybrid system only allow identification of genes that encode polypeptides that directly interact with a known polypeptide. Many of the currently available methods for identifying genes that function in a particular pathway require cumbersome analysis of complex phenotypes. Consequently, in many of the available methods, identification of the genes that function in the pathway of interest is neither rapid nor efficient. Methods that are capable of identifying the underlying genes that regulate important biological pathways, such as the plant pathogenesis response or mammalian tumor progression, would thus be of great value.
Clearly a general method of functional genetic analysis is needed. The method should be simple, rapid, allow high throughput screening, and permit identification of components of genetic pathways that regulate traits of interest. The method should not require an understanding of the detailed basis of a particular phenotype or the mechanisms that underlie specific cellular behaviors. The method should be generally applicable to a great variety of cells, including cells cultured from somatic tissues of multicellular organisms, and it should allow rapid isolation of the nucleotide sequences that function in the pathway of interest.